American Economic Review · 2026 · Daniel Björkegren、Blumenstock Joshua E.、Samsun Knight
中文摘要
越来越多的决策由机器学习算法指导。然而，当具有重要后果的决策被编码进算法时，个体可能会策略性地改变其行为以实现期望的结果。本文开发了一种实证方法，对决策算法进行调整以预判操纵行为。通过显式地对操纵激励进行建模，我们的方法所生成的决策规则在操纵下保持稳定，即便这些规则完全透明也是如此。我们通过在肯尼亚开展的一项大规模实地实验对该方法进行压力测试。在实施时，线性策略稳健决策规则的表现优于LASSO等标准线性模型。(JEL C45, C93, D12, D91, G51, O12, O16)
Abstract
An increasing number of decisions are guided by machine learning algorithms. But when consequential decisions are encoded in algorithms, individuals may strategically alter their behavior to achieve desired outcomes. This paper develops an empirical approach that adjusts decision algorithms to anticipate manipulation. By explicitly modeling incentives to manipulate, our approach produces decision rules that are stable under manipulation, even when the rules are fully transparent. We stress-test this approach through a large field experiment in Kenya. When implemented, linear strategy-robust decision rules outperform standard linear models such as LASSO. (JEL C45, C93, D12, D91, G51, O12, O16)